把csv入库到hadoop集群

来源：互联网发布：kmp算法学不会怎么办编辑：程序博客网时间：2024/05/01 06:59

--建立文本表

create table tmp.S_CONF_GRIDDING_LST_text(
s_id varchar(160)
,s_date timestamp
)ROW FORMAT DELIMITED FIELDS TERMINATED BY ',' STORED AS TEXTFILE ;

--拷贝到hdfs

[root@MASTER01 ~]# hadoop fs -copyFromLocal /root/008.csv /user/hive/warehouse/tmp.db/
[root@MASTER01 ~]# hadoop fs -ls hdfs://MASTER01:8020/user/hive/warehouse/tmp.db/008.csv
-rw-r--r-- 3 root hive 48 2016-04-15 12:53 hdfs://MASTER01:8020/user/hive/warehouse/tmp.db/008.csv
[root@MASTER01 ~]#

--load到文本表

LOAD DATA INPATH '/user/hive/warehouse/tmp.db/008.csv' OVERWRITE INTO TABLE S_CONF_GRIDDING_LST_text;
Query: load DATA INPATH '/user/hive/warehouse/tmp.db/008.csv' OVERWRITE INTO TABLE S_CONF_GRIDDING_LST_text
+----------------------------------------------------------+
| summary |
+----------------------------------------------------------+
| Loaded 1 file(s). Total files in destination location: 1 |
+----------------------------------------------------------+
Fetched 1 row(s) in 0.11s

--建立正式表
[slave02:21000] > create table tmp.S_CONF_GRIDDING_LST(
> s_id varchar(160)
> ,s_date timestamp
> )stored as parquet;

--把文本表放入正式表
insert into S_CONF_GRIDDING_LST select * from S_CONF_GRIDDING_LST_text;

0 0